Introduction: PRICAI 2016 is the International Conference on Artificial Intelligence at the Pacific Rim and is held every two years. The conference focuses on the theory of artificial intelligence, technology and its application in the social field, and its importance to the economies of Pacific Rim countries.
Instance Selection Method for Improving Graph-Based Semi-supervised Learning 
Abstract: Graph-based semi-supervised learning (GSSL) is one of the most important semi-supervised learning (SSL) paradigms. Although the GSSL method is useful in many situations, it may degrade performance when using untagged data. In this article, we propose a new GSSL approach: GSSLIS based on case selection to reduce the possibility of performance degradation. Our basic idea is to give a series of unlabeled instances. Using all unlabeled instances is not the best approach; instead, we should use unlabeled instances that have a high probability of helping to improve performance, while avoiding the use of There are high risk unlabeled instances. A large number of data experiments have shown that the chance that our proposed method deteriorates is much less than most optimal GSSL methods.
Keywords: semi-supervised learning based on performance degradation · · instance of the selected plot
First author introduction
Hai Wang
Position: MSc, Department of Computer Science and Technology, Nanjing University / LAMDA Group
Research direction: Data Mining, Machine Learning
Instructor: Li Yufeng
Position: Ph.D. Assistant Professor, Department of Computer Science and Technology, Nanjing University/LAMDA Group
Research direction: machine learning, data mining, semi-supervised learning, multi-instance learning, multi-label learning, etc.
Related publications:
·Graph Quality Judgement: A Large Margin Expedition(IJCAI,2016)
·Kenerlized matrix factorization for collaborative filtering(SDM,2016)
Via:PRICAI 2016
PS : This article was compiled by Lei Feng Network (search “Lei Feng Network†public number) and it was compiled without permission.
Original paper download